1perf-top(1) 2=========== 3 4NAME 5---- 6perf-top - System profiling tool. 7 8SYNOPSIS 9-------- 10[verse] 11'perf top' [-e <EVENT> | --event=EVENT] [<options>] 12 13DESCRIPTION 14----------- 15This command generates and displays a performance counter profile in real time. 16 17 18OPTIONS 19------- 20-a:: 21--all-cpus:: 22 System-wide collection. (default) 23 24-c <count>:: 25--count=<count>:: 26 Event period to sample. 27 28-C <cpu-list>:: 29--cpu=<cpu>:: 30Monitor only on the list of CPUs provided. Multiple CPUs can be provided as a 31comma-separated list with no space: 0,1. Ranges of CPUs are specified with -: 0-2. 32Default is to monitor all CPUS. 33 34-d <seconds>:: 35--delay=<seconds>:: 36 Number of seconds to delay between refreshes. 37 38-e <event>:: 39--event=<event>:: 40 Select the PMU event. Selection can be a symbolic event name 41 (use 'perf list' to list all events) or a raw PMU event in the form 42 of rN where N is a hexadecimal value that represents the raw register 43 encoding with the layout of the event control registers as described 44 by entries in /sys/bus/event_source/devices/cpu/format/*. 45 46-E <entries>:: 47--entries=<entries>:: 48 Display this many functions. 49 50-f <count>:: 51--count-filter=<count>:: 52 Only display functions with more events than this. 53 54--group-sort-idx:: 55 Sort the output by the event at the index n in group. If n is invalid, 56 sort by the first event. It can support multiple groups with different 57 amount of events. WARNING: This should be used on grouped events. 58 59-F <freq>:: 60--freq=<freq>:: 61 Profile at this frequency. Use 'max' to use the currently maximum 62 allowed frequency, i.e. the value in the kernel.perf_event_max_sample_rate 63 sysctl. 64 65-i:: 66--inherit:: 67 Child tasks do not inherit counters. 68 69-k <path>:: 70--vmlinux=<path>:: 71 Path to vmlinux. Required for annotation functionality. 72 73--ignore-vmlinux:: 74 Ignore vmlinux files. 75 76--kallsyms=<file>:: 77 kallsyms pathname 78 79-m <pages>:: 80--mmap-pages=<pages>:: 81 Number of mmap data pages (must be a power of two) or size 82 specification with appended unit character - B/K/M/G. The 83 size is rounded up to have nearest pages power of two value. 84 85-p <pid>:: 86--pid=<pid>:: 87 Profile events on existing Process ID (comma separated list). 88 89-t <tid>:: 90--tid=<tid>:: 91 Profile events on existing thread ID (comma separated list). 92 93-u:: 94--uid=:: 95 Record events in threads owned by uid. Name or number. 96 97-r <priority>:: 98--realtime=<priority>:: 99 Collect data with this RT SCHED_FIFO priority. 100 101--sym-annotate=<symbol>:: 102 Annotate this symbol. 103 104-K:: 105--hide_kernel_symbols:: 106 Hide kernel symbols. 107 108-U:: 109--hide_user_symbols:: 110 Hide user symbols. 111 112--demangle-kernel:: 113 Demangle kernel symbols. 114 115-D:: 116--dump-symtab:: 117 Dump the symbol table used for profiling. 118 119-v:: 120--verbose:: 121 Be more verbose (show counter open errors, etc). 122 123-z:: 124--zero:: 125 Zero history across display updates. 126 127-s:: 128--sort:: 129 Sort by key(s): pid, comm, dso, symbol, parent, srcline, weight, 130 local_weight, abort, in_tx, transaction, overhead, sample, period. 131 Please see description of --sort in the perf-report man page. 132 133--fields=:: 134 Specify output field - multiple keys can be specified in CSV format. 135 Following fields are available: 136 overhead, overhead_sys, overhead_us, overhead_children, sample and period. 137 Also it can contain any sort key(s). 138 139 By default, every sort keys not specified in --field will be appended 140 automatically. 141 142-n:: 143--show-nr-samples:: 144 Show a column with the number of samples. 145 146--show-total-period:: 147 Show a column with the sum of periods. 148 149--dsos:: 150 Only consider symbols in these dsos. This option will affect the 151 percentage of the overhead column. See --percentage for more info. 152 153--comms:: 154 Only consider symbols in these comms. This option will affect the 155 percentage of the overhead column. See --percentage for more info. 156 157--symbols:: 158 Only consider these symbols. This option will affect the 159 percentage of the overhead column. See --percentage for more info. 160 161-M:: 162--disassembler-style=:: Set disassembler style for objdump. 163 164--prefix=PREFIX:: 165--prefix-strip=N:: 166 Remove first N entries from source file path names in executables 167 and add PREFIX. This allows to display source code compiled on systems 168 with different file system layout. 169 170--source:: 171 Interleave source code with assembly code. Enabled by default, 172 disable with --no-source. 173 174--asm-raw:: 175 Show raw instruction encoding of assembly instructions. 176 177-g:: 178 Enables call-graph (stack chain/backtrace) recording. 179 180--call-graph [mode,type,min[,limit],order[,key][,branch]]:: 181 Setup and enable call-graph (stack chain/backtrace) recording, 182 implies -g. See `--call-graph` section in perf-record and 183 perf-report man pages for details. 184 185--children:: 186 Accumulate callchain of children to parent entry so that then can 187 show up in the output. The output will have a new "Children" column 188 and will be sorted on the data. It requires -g/--call-graph option 189 enabled. See the `overhead calculation' section for more details. 190 Enabled by default, disable with --no-children. 191 192--max-stack:: 193 Set the stack depth limit when parsing the callchain, anything 194 beyond the specified depth will be ignored. This is a trade-off 195 between information loss and faster processing especially for 196 workloads that can have a very long callchain stack. 197 198 Default: /proc/sys/kernel/perf_event_max_stack when present, 127 otherwise. 199 200--ignore-callees=<regex>:: 201 Ignore callees of the function(s) matching the given regex. 202 This has the effect of collecting the callers of each such 203 function into one place in the call-graph tree. 204 205--percent-limit:: 206 Do not show entries which have an overhead under that percent. 207 (Default: 0). 208 209--percentage:: 210 Determine how to display the overhead percentage of filtered entries. 211 Filters can be applied by --comms, --dsos and/or --symbols options and 212 Zoom operations on the TUI (thread, dso, etc). 213 214 "relative" means it's relative to filtered entries only so that the 215 sum of shown entries will be always 100%. "absolute" means it retains 216 the original value before and after the filter is applied. 217 218-w:: 219--column-widths=<width[,width...]>:: 220 Force each column width to the provided list, for large terminal 221 readability. 0 means no limit (default behavior). 222 223--proc-map-timeout:: 224 When processing pre-existing threads /proc/XXX/mmap, it may take 225 a long time, because the file may be huge. A time out is needed 226 in such cases. 227 This option sets the time out limit. The default value is 500 ms. 228 229 230-b:: 231--branch-any:: 232 Enable taken branch stack sampling. Any type of taken branch may be sampled. 233 This is a shortcut for --branch-filter any. See --branch-filter for more infos. 234 235-j:: 236--branch-filter:: 237 Enable taken branch stack sampling. Each sample captures a series of consecutive 238 taken branches. The number of branches captured with each sample depends on the 239 underlying hardware, the type of branches of interest, and the executed code. 240 It is possible to select the types of branches captured by enabling filters. 241 For a full list of modifiers please see the perf record manpage. 242 243 The option requires at least one branch type among any, any_call, any_ret, ind_call, cond. 244 The privilege levels may be omitted, in which case, the privilege levels of the associated 245 event are applied to the branch filter. Both kernel (k) and hypervisor (hv) privilege 246 levels are subject to permissions. When sampling on multiple events, branch stack sampling 247 is enabled for all the sampling events. The sampled branch type is the same for all events. 248 The various filters must be specified as a comma separated list: --branch-filter any_ret,u,k 249 Note that this feature may not be available on all processors. 250 251--raw-trace:: 252 When displaying traceevent output, do not use print fmt or plugins. 253 254--hierarchy:: 255 Enable hierarchy output. 256 257--overwrite:: 258 Enable this to use just the most recent records, which helps in high core count 259 machines such as Knights Landing/Mill, but right now is disabled by default as 260 the pausing used in this technique is leading to loss of metadata events such 261 as PERF_RECORD_MMAP which makes 'perf top' unable to resolve samples, leading 262 to lots of unknown samples appearing on the UI. Enable this if you are in such 263 machines and profiling a workload that doesn't creates short lived threads and/or 264 doesn't uses many executable mmap operations. Work is being planed to solve 265 this situation, till then, this will remain disabled by default. 266 267--force:: 268 Don't do ownership validation. 269 270--num-thread-synthesize:: 271 The number of threads to run when synthesizing events for existing processes. 272 By default, the number of threads equals to the number of online CPUs. 273 274--namespaces:: 275 Record events of type PERF_RECORD_NAMESPACES and display it with the 276 'cgroup_id' sort key. 277 278-G name:: 279--cgroup name:: 280monitor only in the container (cgroup) called "name". This option is available only 281in per-cpu mode. The cgroup filesystem must be mounted. All threads belonging to 282container "name" are monitored when they run on the monitored CPUs. Multiple cgroups 283can be provided. Each cgroup is applied to the corresponding event, i.e., first cgroup 284to first event, second cgroup to second event and so on. It is possible to provide 285an empty cgroup (monitor all the time) using, e.g., -G foo,,bar. Cgroups must have 286corresponding events, i.e., they always refer to events defined earlier on the command 287line. If the user wants to track multiple events for a specific cgroup, the user can 288use '-e e1 -e e2 -G foo,foo' or just use '-e e1 -e e2 -G foo'. 289 290--all-cgroups:: 291 Record events of type PERF_RECORD_CGROUP and display it with the 292 'cgroup' sort key. 293 294--switch-on EVENT_NAME:: 295 Only consider events after this event is found. 296 297 E.g.: 298 299 Find out where broadcast packets are handled 300 301 perf probe -L icmp_rcv 302 303 Insert a probe there: 304 305 perf probe icmp_rcv:59 306 307 Start perf top and ask it to only consider the cycles events when a 308 broadcast packet arrives This will show a menu with two entries and 309 will start counting when a broadcast packet arrives: 310 311 perf top -e cycles,probe:icmp_rcv --switch-on=probe:icmp_rcv 312 313 Alternatively one can ask for a group and then two overhead columns 314 will appear, the first for cycles and the second for the switch-on event. 315 316 perf top -e '{cycles,probe:icmp_rcv}' --switch-on=probe:icmp_rcv 317 318 This may be interesting to measure a workload only after some initialization 319 phase is over, i.e. insert a perf probe at that point and use the above 320 examples replacing probe:icmp_rcv with the just-after-init probe. 321 322--switch-off EVENT_NAME:: 323 Stop considering events after this event is found. 324 325--show-on-off-events:: 326 Show the --switch-on/off events too. This has no effect in 'perf top' now 327 but probably we'll make the default not to show the switch-on/off events 328 on the --group mode and if there is only one event besides the off/on ones, 329 go straight to the histogram browser, just like 'perf top' with no events 330 explicitly specified does. 331 332--stitch-lbr:: 333 Show callgraph with stitched LBRs, which may have more complete 334 callgraph. The option must be used with --call-graph lbr recording. 335 Disabled by default. In common cases with call stack overflows, 336 it can recreate better call stacks than the default lbr call stack 337 output. But this approach is not full proof. There can be cases 338 where it creates incorrect call stacks from incorrect matches. 339 The known limitations include exception handing such as 340 setjmp/longjmp will have calls/returns not match. 341 342ifdef::HAVE_LIBPFM[] 343--pfm-events events:: 344Select a PMU event using libpfm4 syntax (see http://perfmon2.sf.net) 345including support for event filters. For example '--pfm-events 346inst_retired:any_p:u:c=1:i'. More than one event can be passed to the 347option using the comma separator. Hardware events and generic hardware 348events cannot be mixed together. The latter must be used with the -e 349option. The -e option and this one can be mixed and matched. Events 350can be grouped using the {} notation. 351endif::HAVE_LIBPFM[] 352 353INTERACTIVE PROMPTING KEYS 354-------------------------- 355 356[d]:: 357 Display refresh delay. 358 359[e]:: 360 Number of entries to display. 361 362[E]:: 363 Event to display when multiple counters are active. 364 365[f]:: 366 Profile display filter (>= hit count). 367 368[F]:: 369 Annotation display filter (>= % of total). 370 371[s]:: 372 Annotate symbol. 373 374[S]:: 375 Stop annotation, return to full profile display. 376 377[K]:: 378 Hide kernel symbols. 379 380[U]:: 381 Hide user symbols. 382 383[z]:: 384 Toggle event count zeroing across display updates. 385 386[qQ]:: 387 Quit. 388 389Pressing any unmapped key displays a menu, and prompts for input. 390 391include::callchain-overhead-calculation.txt[] 392 393SEE ALSO 394-------- 395linkperf:perf-stat[1], linkperf:perf-list[1], linkperf:perf-report[1] 396